Skip to contents

Derive ROC-based biomarker cutoffs using adjusted risk

Source: R/derive_cutoffs_sensspec.R
derive_cutoffs_sensspec.Rd

Implements a ROC-guided two-cutoff approach to classify subjects into rule-out, rule-in, or indeterminate zones using confounder-adjusted risk scores.

Usage

derive_cutoffs_sensspec(
  adjust_result,
  biomarker,
  outcome,
  sens_target = 0.8,
  spec_target = 0.8,
  loess_span = 0.35
)

Arguments

adjust_result

Output list from adjust_score(), containing the augmented dataset with biomarker, adjusted_risk, and outcome.

biomarker

Character, name of biomarker column.

outcome

Character, name of binary disease outcome column (0/1). If missing, the outcome stored in adjust_result will be used.

sens_target

Numeric, target sensitivity for rule-out threshold (default = 0.80).

spec_target

Numeric, target specificity for rule-in threshold (default = 0.80).

loess_span

Numeric, smoothing parameter for LOESS regression (default = 0.35). Smaller values give more flexibility, larger values give smoother fits.

Value

List with rule-out cutoff (c_L), rule-in cutoff (c_H), classification vector, and augmented dataset.

Details

Method

  • Compute ROC curve with predictor = adjusted_risk and outcome = true disease status.

  • Identify thresholds:

    • Rule-out threshold (T_L): lowest risk threshold with sensitivity\(\geq\) sens_target (default 0.80).

    • Rule-in threshold (T_H): highest risk threshold with specificity\(\geq\) spec_target (default 0.80).

  • Map thresholds back to biomarker scale by fitting a LOESS regression (adjusted_risk ~ biomarker, default span = 0.35 (user-adjustable via loess_span)) and interpolating biomarker values at T_L and T_H.

  • Classification:

    • biomarker < c_L= "Negative" (rule-out).

    • biomarker > c_H= "Positive" (rule-in).

    • otherwise = "Indeterminate".

References

Giacomucci, G., Crucitti, C., Ingannato, A., et al. (2025). The two cut-offs approach for plasma p-tau217 in detecting Alzheimer's disease. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 17:e70116.

Hazan, J., Liu, K.Y., Isaacs, J.D., Howard, R. (2025). Cut-points and gray zones: The challenges of integrating Alzheimer's disease plasma biomarkers into clinical practice. Alzheimer's & Dementia, 21:e70113.

Author

Bhrigu Kumar Rajbongshi, Seyed Ehsan Saffari and Nastaran Marzban

Examples

n <- 500
set.seed(123)

# outcome
DISEASE <- rbinom(n, 1, 0.35)

# biomarker depends on disease
PTAU <- rnorm(n, mean = 25 + 10 * DISEASE, sd = 8)

# confounders (optional noise)
AGE <- round(rnorm(n, 73, 7))
SEX <- sample(c(0,1), n, replace = TRUE)
BMI <- rnorm(n, 25, 4)
HTN <- rbinom(n, 1, 0.3)
DM <- rbinom(n, 1, 0.2)

df <- data.frame(PTAU, AGE, SEX, BMI, HTN, DM, DISEASE)

res <- adjust_score(
 df,
 biomarker = "PTAU",
 outcome = "DISEASE",
 confounders = c("AGE","SEX","BMI","HTN","DM")
)
#> number of iterations= 704 
#> Setting levels: control = 0, case = 1
#> Setting direction: controls < cases

cutoffs <- derive_cutoffs_sensspec(res, biomarker = "PTAU", outcome = "DISEASE")
#> Warning: Sensitivity target not achievable -> using 5th percentile of adjusted risk as fallback threshold.
#> Warning: Specificity target not achievable -> using 95th percentile of adjusted risk as fallback threshold.
cutoffs$c_L; cutoffs$c_H
#> [1] 9.467853
#> [1] 44.06341
table(cutoffs$class)
#> 
#> Indeterminate      Negative      Positive 
#>           461            13            26